In April, just a few weeks after the spring semester pivoted unceremoniously to digital, Catherine Klapperich, a biomedical engineering professor at Boston University, was thinking about the fall. The Boston area had been inundated with Covid-19 cases, and at the time, tests remained scant. But the university had come to her with an unfathomable question: In four months’ time, how would they test students and staff when they returned to campus? The university didn’t have its own testing lab. So Klapperich, who studies medical diagnostics, was tasked with designing one.
“We had an empty room,” Klapperich says. “Actually, we didn’t even have an empty room. We had a room we had to empty.” Her team worked to repurpose the space, in both physical and bureaucratic terms. That meant getting the right clinical licenses to perform diagnostics tests and deliver results, plus trained staff to do it. And then filling the room with a battery of robotic instruments, sequencing machines, reagents, and nasal swabs. With all that, Klapperich hopes, the university will begin testing faculty, staff, and students on a regular basis next month, with capacity for roughly 5,000 tests per day.
That testing is at the center of a strategy to aggressively monitor the school for outbreaks, with “isolation dorms” on a remote patch of campus for anyone who gets sick, contact tracing staff, and apps to let students report symptoms and stay up to date on tests. Universities have become a microcosm of the pandemic strategy the government has largely failed to enact. “We’re all the federal government now,” Klapperich says.
Since the beginning of the pandemic, public health advocates have boiled down virus containment to a few simple steps. First, get the virus under control using the blunt tools of isolation and social distancing. And then, as those measures soften: test, trace, isolate. The process of quickly identifying exposed people, coupled with behaviors that are by now common sense, like wearing masks, washing hands, and limiting large gatherings, would keep the virus’s spread, at the very least, to a much slower creep.
We all know how that’s gone. Six months into the pandemic, few places in the US have the virus under control, with the Northeast (for now) one of the major exceptions. But next month, hundreds of thousands of college students will return to campus all the same, often in places where outbreaks are actively raging, or could soon spark. Universities must keep the pandemic at bay, all while trying to maintain for students some of the benefits of being physically on campus: to socialize, to work in labs, to participate in events and activities. All the things that classes via Zoom don’t permit.
“We cannot keep them locked up in their dorms. There’s no point,” says Amy Gorin, a behavioral psychologist at the University of Connecticut involved in reopening efforts there. “So how do we approximate the normal experience without compromising on safety?”
Large research universities have a leg up in offering tests. They can create pop-up labs that draw from existing resources, as schools like Boston University and UC Berkeley have done, or look to affiliated hospitals for help. Some smaller campuses have taken a collaborative approach. A number of colleges across New England, including Wellesley, Colby, and Williams, recently signed on to send test samples to the Broad Institute, a research center affiliated with MIT and Harvard that has opened its high-throughput Covid-19 lab to other educational institutions. Others must compete for capacity at commercial labs.
But far more universities have instead opted against surveillance testing, reserving testing for students who show symptoms or have a known risk of exposure. That’s in line with current guidance from the Centers for Disease Control and Prevention, which says there is not enough evidence from testing on college campuses to determine whether it is useful or not.
Residential college campuses may appear, in some ways, to be well-suited to handling the pandemic. They’re filled with young people who are less likely to experience severe cases of the disease, and who won’t bring their germs home each afternoon to mom and dad. But in other ways, they’re porous, says David Paltiel, a professor of public health policy at Yale University. Even in the sluggish, depopulated days of summer, a number of universities have seen outbreaks in places like frat houses and sports practices. Plus, the risks are uneven. If the virus moves fast within a population of college students, it’s the older professors, janitors and dining hall workers who bear the risk. “We all worry about the safety of students, but the cost of a mistake is likely to be borne by the most vulnerable members of the community,” he says.
There’s no guarantee that even the most rigorous testing strategies will prevent outbreaks. That will depend on the actual process going off without a hitch—no easy task with thousands of students and staff. And tests themselves don’t prevent infection. That requires mask-wearing, decontamination, and social distancing on campus. In some places, such as the Georgia public university system, professors have been fighting for those basic protections.
Some of the campuses that have planned surveillance testing, such as Harvard and MIT, are also making a partial shift to coursework online, reducing the number of students and faculty on campus. But the details are fraught. A number of colleges have given professors the choice of virtual or in-person teaching, others have not. Boston University requires faculty to ask for an exemption to in-person, a move that has drawn the ire of teaching staff there. More than half of US colleges and universities intend to hold classes fully in person, according to a Chronicle of Higher Education survey updated Wednesday, and 30 percent plan a hybrid model involving some online instruction.
For those schools, frequent testing should be a basic requirement, Paltiel argues. Without regular testing, outbreaks can easily fester and spread out of control before a symptomatic person is ever identified. The dynamic played out in the early days of the pandemic, when asymptomatic cases from abroad slipped by symptom-based screening at airports, seeding community spread.
In a modeling study that has been accepted for publication, but not yet appeared in peer-reviewed form, Paltiel and Rochelle Walensky, chief of infectious disease at Massachusetts General Hospital, gamed out what theoretical outbreaks might look like at a college of about 5,000 students. They seeded their models with small numbers of cases in an otherwise healthy student population, balancing the costs of various testing strategies against the costs of potential infections that go undetected.
Their model showed that if testing was done frequently enough—they found once every three days was optimal—universities could manage with cheaper testing methods that miss as many as a third of infections, because an individual patient is unlikely to have multiple false negatives in a row. “Frequency of testing is the most powerful variable that the administrators of a university control. That’s what really matters here,” Paltiel says. That would also reduce the burden on contact tracing efforts, he notes, since infected contacts will be identified in subsequent rounds of testing.
The researchers also found that less-frequent testing—let alone testing only symptomatic people—meant outbreaks swiftly got out of hand. “We looked at a lot of scenarios and didn’t find a single one under which that would be sufficient to contain an outbreak,” Paltiel says. “Most universities are thinking in terms of what they can afford, not what would actually work.”
Beyond the costs of setting up a clinical lab, a diagnostic test for Covid-19 is typically billed to insurance at about $100. In-house testing can be cheaper (the Broad Institute says it plans to offer tests for $25 to $30 each) but an additional headwind is that tests for asymptomatic people are not typically covered by health plans. In other words, universities will foot the bill. “For now, we’ve gone with the fundamental assumption that we will not get reimbursed,” Klapperich says.
“I’m painfully aware that what we’re recommending may be beyond the reach of many, if not most, of the universities in the country,” Paltiel says. “But if you can’t see your way logistically or financially towards implementing this strategy, then you should be asking yourself if you have any business reopening.”
It’s possible that other, cheaper testing methods could eventually change that calculus. The FDA is expected to issue guidance soon on pooled testing, for example, which would allow labs to mix samples in batches. That’s a useful strategy provided transmission rates are low, because if all tests in a pool are negative, labs can avoid testing them individually. Dozens of cheaper tests designed for surveillance are also under regulatory consideration. But those will likely have to wait for future semesters, with the fall session only weeks away and universities already scrambling to ready complex procedures to gather samples from students, coordinate move-ins, and figure out who will give the tests. “If only we had eight or nine weeks to figure this out, not three or four,” Paltiel says.
In the real world, repeat testing also requires an infrastructure that works smoothly. At the University of Connecticut, which plans to test a percentage of the student body each week, Gorin and her team of behavioral scientists have been developing ways to get students comfortable with the trials of repeat testing, contact tracing—and the looming possibility of being sent to an isolation dorm if they or a close contact test positive.
Gorin usually works in obesity research—studying how motivation and environmental factors affect a person’s ability to lose weight. There are similarities in guiding people to reduce the spread of Covid-19, she says. In interviews and surveys with students, her team found that they were on board with the big-picture goals. “Their motivation is to keep campus open and to not make somebody else sick,” she says. “Students want to make this work.” But the challenge comes from the gritty details of life back on campus, which provides constant temptations to revert to normalcy.
Those temptations will likely begin during a two-week quarantine at the beginning of the semester, which school officials have planned to complete initial rounds of testing and await results. Students suggested to the researchers that the quarantine would quickly crumble. They also pointed to the potential shame of testing positive and worries about contact tracing, suggesting their peers might be reluctant to share information with university staff. “People are concerned about causing a hundred of their friends to be in quarantine,” Gorin says. Others expressed concerns about privacy. Say they’d recently been to a bar; would college staff reprimand them for underage drinking?
Gorin’s team has advocated taking a harm reduction approach. “You know people are going to engage in some behaviors that will put them at risk. What is the safest way for them to be together socially?” she asks. “Better to be outside, to make sure everyone has their own cup.” They’re considering prizes for using symptom tracking apps, and suggest the university plan social activities outdoors, where the risk of transmission is lower, and to keep people from going stir-crazy. (Rice University, citing Houston’s year-round balmy weather, announced Wednesday it had upped the ante, opting to hold classes in tents.)
But it’s impossible to predict what will happen, Gorin notes. Setting aside student behavior, the virus is, ultimately, out of any one university’s control. While school administrators in New England, where infection rates are currently low, may feel optimistic about tests being readily available to the community, it’s unclear what will happen if cases spike once again. Places with once abundant testing capacity, like the San Francisco Bay Area, now have week-long waits for appointments to get swabbed.
For now, Gorin says the university is thinking short term—it’s “a cautious reopening,” she says. The semester ends at Thanksgiving break. After that, it’s unclear when they’ll be coming back. That will depend on what transpires in the middle of flu season and a New England winter that makes outdoor activities difficult. “I think we’re all approaching this with apprehension,” Gorin says. “We’re hopeful we can get through the fall. Spring is another question.”
- How masks went from don’t-wear to must-have
- Q&A: Larry Brilliant on how well we are fighting Covid-19
- Covid-19 is accelerating human transformation—let’s not waste it
- 15 face masks we actually like to wear
- After the virus: How we'll learn, age, move, listen, and create
- Read all of our coronavirus coverage here
